Alcoholism in women represents a major and increasing public health problem. Previous genetic and genetic-epidemiologic studies of alcoholism have focused largely on males, and almost exclusively on treated samples. The first two years of this grant supported analyses of previously collected data on 2,163 personally interviewed twins from female-female pairs from the population-based Virginia Twin Registry and 1,470 of their surviving parents. These analyses yielded exciting results suggesting that genetic factors play a major role in the etiology of alcoholism in women. However, because the major focus of previous interviews in this sample was on depression and anxiety, the clinical information collected on alcoholism is quite limited. Therefore, this competitive renewal requests support for a follow-up telephone interview of all members of this unique cohort of female-female twins. This re-assessment, conducted 7 years after their initial personal interview, will focus specifically on broadly defined alcoholism and alcohol-related risk factors. This proposal has 6 major aims the first of which is to replicate and extend our previous findings that in women genetic factors play a major etiologic role in alcoholism. Statistical power will be increased by new onsets, increased reliability due to repeat assessments and a substantial expansion of the range of alcohol-related problems that will be examined. Other aims include: i) to determine whether the discrepancy between our initial results on the role of genetic factors in alcoholism in women and clinically ascertained female twin samples results from differences in sample selection, ii) understand more fully the causes of comorbidity in women between alcoholism and mood and anxiety disorders, iii) clarify the inter-relationship between personality and liability to alcoholism in women, iv) examine the inter-relationship between the genetic and environmental factors in women which influence the probability of lifetime abstinence from alcohol and those which influence the risk for alcoholism given non-abstinence, and v) to use our longitudinal and twin-family design, including the examination of offspring of parents with alcoholism as determined by personal interview, to uncover clinical and etiologic heterogeneity in alcoholism in women. The first two years of this proposal focus on data collection and the final two on data analysis. Data analysis will include state-of-the-art statistical and genetic methods.